Custom timestamp field off by timezone offset

Elastic Stack Logstash
1

I'm working on revamping some indexes for data and am running into an issue with times. I'm currently ingesting entries like this one below:

<164>Nov 09 2017 11:26:09: %ASA-4-106023: Deny tcp src outside:191.101.167.235/45269 dst outside:65.126.243.209/8545 by access-group "private_in" [0x0, 0x0]

Using the cisco filters all the data is parsing out correctly and going into the correct fields. What I would like to do is use the field eventtime as the time filter field name. In the input filter I'm using the following statement to match the time (correctly):

date {
match => ["timestamp","MMM dd yyyy HH:mm:ss","MMM d yyyy HH:mm:ss"]
}

Since I'm using a custom index pattern I've created a mapping for the timestamp field:
"timestamp" : {"type" : "date", "format" : "MMM dd YYYY hh:mm:ss"},

Once the time appears in the index template at this point the time changes to November 9th 2017, 05:26:09.000

If I let the field be created automatically the time is recognized properly, however, I can't use the field as the time filter field. Would there be a way to just append the timezone to the timestamp field for this specific filter?

Thanks!

2

date {
match => ["timestamp","MMM dd yyyy HH:mm:ss","MMM d yyyy HH:mm:ss"]
}

Since I'm using a custom index pattern I've created a mapping for the timestamp field:
"timestamp" : {"type" : "date", "format" : "MMM dd YYYY hh:mm:ss"},

I'm confused. You're using the date filter to parse timestamp and store in @timestamp, but you're also storing the raw timestamp value and configuring ES to recognize the timestamp format. Why do you want to store the same timestamp in different fields and in different formats?

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.